Against Treating Introspection As Perception-Like

Psyche, Volume 16, number 1

Against Treating Introspection as Perception-Like

Abstract

A perceptual theory of introspection is one that treats introspection as a species of perception or as a special case of perception. Additionally, a perceptual theory of introspection is one for which introspection shares at least some of the essential features of perception. However, I will show that there are certain essential features of perception that introspection lacks. Moreover, those features common to perception and introspection are insufficient to distinguish perception from belief. Thus, there is good reason to deny that introspection fits a perceptual model of introspection.

Renee J Smith1 Associate Professor of Philosophy Department of Philosophy and Religious Studies Coastal Carolina University

A perceptual theory of introspection is one that treats introspection as a species of perception or as a special case of perception. While introspection and perception are dissimilar in certain respects, a perceptual theory of introspection will be based on certain fundamental similarities between them. What I take to be an essential feature of any perceptual theory of introspection is that it takes perception to be a basic mental state, one irreducible to some other mental state. This feature rules out treating perception as a species of belief (what Dretske has called a cognitivist theory of perception) and then likening introspection to perception on the basis that it, too, is a species of belief. This latter view of introspection would be classified as a non-perceptual theory of introspection because while it does reduce introspection to perception, it goes on to reduce perception to belief. A perceptual theory of introspection, in addition to taking perception to be a basic mental state, is one for which introspection shares at least some of the essential features of perception. I will show, however, that there are certain essential features of perception that introspection lacks. In addition, those features common to perception and introspection are insufficient to distinguish perception from belief. Thus, there is good reason to deny that introspection fits a perceptual model of introspection.

1Conway SC 29528-6054 843-349-2083 [email protected] http://ww2.coastal.edu/rsmith

Psyche, Volume 16, number 1 Against treating introspection

Perception, in the most straightforward case, is seeing, hearing, tasting, smelling, and touching; namely, it is sense perception. It is acquiring information about our surroundings via our sense organs. Certainly, when we introspect, we do not see, hear, taste, smell or touch our mental states; so, the employment of the sense organs cannot ground a perceptual theory of introspection. Other cases of perception, however, do not fit these stereotypical cases either, for example, proprioception.2 It seems that one can be (perceptually) aware that one’s leg is bent without seeing or touching it, and yet this usually counts as a case of perception. So, rather than thinking of perception as dependent on the sense organs, we might simply think of it as a means of acquiring information about our environment (including what might be called our somatic environment or our non-mental environment.) Since introspection, let us say, is directed inward, and perception seems to be directed outward, what is common to both is that we acquire information about whatever is perceived or introspected. By perceiving something, one stands in a particular relation to the thing seen. Assume that this is a causal relation between the perceptual object and the perceiver.3 So, in order to see a cat, one must be causally related to certain intrinsic properties of the cat. One acquires information about the cat (its location, color, size, and shape) in virtue of standing in this relation to a cat. If cat experiences are produced by something other than cats, then they do not count as perceptual experiences (of cats). This information is then available to be (or to enter into) the content of perceptual belief. Seeing the cat allows one to form the belief that what one sees is a cat. The ability to form perceptual beliefs will be dependent upon possessing certain concepts (e.g. the concept CAT); simply seeing the cat, standing in the perceptual relation to the cat, however, does not seem to be concept-dependent. One can be aware of the cat without being aware that it is a cat. So we should distinguish simple “seeing” (being aware of the cat) from “perception-that” (being aware that it is a cat).4 The latter, unlike the former, is concept-dependent. One could not see that 5 something is a cat unless one had the concept CAT. Additionally, in the case of simple perception, there is an associated appearance or sensation.6 The cat appears to the perceiver in some way; there is something that it is like to be

2 Also, Bach-y-Rita (Dennett, 1991, pp. 338-343) prosthetic devices, which are described as producing perception- like experiences while bypassing the (relevant) sense organs, do not fit. I am inclined to follow Dennett: This is not a case of seeing, but feeling. 3 (Grice, 1961). It will be one thing to say that the perceptual relation is a causal one and quite another to be committed to Grice’s causal theory of perception (as he takes perception to be the perception of a sense datum caused by a material object). Instead, some have emphasized the acquisition of information that occurs in perception rather than the causal relation. (See Dretske, 2000, 1981). 4 Audi (1988, p. 9) and Dretske (2000) have called this both simple seeing and non-epistemic seeing and contrasted it with epistemic seeing described below as perception-that. Crane (1988, 1992), Tye (1995, 2000), and Peacocke (1992, 2001) all endorse the view that perception has nonconceptual content. McDowell (1994) denies this. 5 For present purposes, I will assume that indeterminate concept possession (see Bealer, 1998) is sufficient for perception-that. That is, one need not know the complete analysis of a concept (or, as Bealer says, have intuitions that each true identity claim for the thing in question is true) in order to possess it. Instead, in possessing the concept CAT one need only have the ability to recognize cats, distinguish them from dogs, tables, chairs, and so forth. (Also see Dretske, 1999, 1995, esp. pp. 59-60, 138-139, 1981). 6 Macdonald (1999) denies that appearances or sensations are essential features of perception (or, she says it is possible that they are not) citing McDowell’s 1994 requirement that all experience be conceptualized. (Appearances being unconceptualized experiences, there are no appearances. It is not clear that Macdonald actually endorses this view.) This cognitivist theory of perception (see Dretske, 2000a, chapter 8) bears certain similarities, in this respect, 80

Psyche, Volume 16, number 1 Against treating introspection aware of the cat even if one is not aware that it is a cat. Simple perception, “awareness-of,” has a certain phenomenal character. Just what this phenomenal character is, or what experience has it, is a central question in philosophy. For the time being, however, we might simply describe it as what is common to both veridical and hallucinatory perceptual experience. What is common to both veridical awareness of a cat and hallucinating a cat is that it seems as if there is a cat one sees. One is having an experience of a cat even if one is not (introspectively) aware that what one is having this experience. How these experiences differ is with respect to the role of the perceptual object. In the case of veridical perception, one stands in a causal relation with a cat. When hallucinating a cat, since there is no cat, it is not the perceptual object that causes one’s cat-awareness. While hallucination (and illusion) are similar with respect to their phenomenal character, and differ with respect to the causal role of the object, we can also distinguish them in terms of the relation they bear to perceptual belief and knowledge. Hallucinating a cat may provide the material for one to come to believe what one sees is a cat, since there is no cat, one cannot know that it is a cat one sees. Therefore, while both veridical and non-veridical perception have phenomenal character and provide the basis for belief, they differ in respect to their knowledge- producing effects. What can be known of non-veridical perception is that it seems as if one sees a cat. How things seem is something that would be acquired by introspection. But is this introspective access sufficiently analogous to perception to warrant adopting a perceptual theory of introspection? There does not appear to be any immediate problem with likening introspection to perception on the basis that it, too, is a causal relation between the subject and the object of introspection. If one introspects on one’s experience of seeing a cat, we could say that the visual experience itself is the cause of one’s introspecting it. Assuming perceptual realism that in order to see a cat, the cat must exist (setting aside cases of non-veridical perception for the moment), we could say that in order to introspect a given mental state, for instance a visual experience, it seems reasonable to say that visual experience must exist.7 There does not seem, however, to be anything analogous to simple perception in introspection. That is, while one can be aware of cats and their properties,8 one cannot be aware of experiences or their properties. Introspecting a visual experience, for example, does not involve that experience appearing in any way.9 While it does seem that there is something that it

to Armstrong’s 1968 theory of perception. However, I think there are good reasons for rejecting a cognitivist theory of perception (again, see Dretske, 2000a), namely, the role simple perception plays in perceptual experience. 7 The natural “objects” (using that term loosely for reasons that will become clear later) of introspection are mental. While I will frequently use perceptual and sensory experiences as examples of states we introspect (for the obvious reason that these are paradigmatically states having phenomenal character, i.e., experiences), it should be understood that cognitive states and emotional states (e.g., beliefs, desires, memories, doubts, fears) are also potential “objects” of introspection. 8 I am not going to further analyze simple perception (beyond object- and property- awareness, see below), but it should be mentioned that there is some debate as to the relation between the two, that is, whether we perceive objects in virtue of perceiving their properties or vice versa. (See Dretske, 2000; Lycan, 1996, particularly chapter 7). 9 Following Lycan (1996), Lyons (1986), Shoemaker (1996), and Searle (1992), I will take this as a datum. Arnold (1997) claims that while there is something that it is like to introspect, it is just different than what it is like to have first-order experiences. This, however, just seems to come out of his commitment to an act-object conception of phenomenal states, a view that has some unsavory consequences I will address later. 81

Psyche, Volume 16, number 1 Against treating introspection is like to see a cat, there is nothing that it is like (over and above what it is like to see the cat) to introspect one’s experience of seeing a cat. Introspection does not seem to have the associated phenomenal character that perception has. Perception involves an awareness of perceptual objects and their properties while introspection does not seem to involve an awareness of (mental) objects and their properties. Moreover, perceptual states involve the thing perceived, appearing to the subject in some way, and states introspected do not appear to the subject in any way. In the case of our introspective awareness of sensations, it may be thought that this is obviously false. Surely, it seems as if we are directly aware of certain non-intentional properties of sensory states, for example, the hurtfulness of pain, pangs of hunger, and so forth. Assuming for the moment that these are non-intentional properties of these states (qualia), and we are introspectively aware of qualia in the same way we are perceptually aware of a cat’s blackness, that is, we simply introspect them: We are p-aware of them. Then presumably, there would be something that it is like to be so aware. The “what it is like” to have an experience (here, a quale) is a property of awareness-of, so if introspection is an awareness-of, then there would be something that it is like to be introspectively aware of qualia. Most people, however, would agree that while pains hurt, there is no additional phenomenal character that results from one’s awareness of one’s pains. Moreover, whereas we can be aware of black even if there is no black object in the vicinity, an analogous case for introspective awareness of pain would entail that we can be aware of a pain quale even if there is no painful state. This suggests something along the lines of sense data that serve as intermediaries in introspective awareness. Thus, there seems to be certain essential features of perception that introspection lacks. A good way to see these differences is within the framework of Dretske’s characterization of different species of awareness.10 He distinguishes two varieties of simple perception: awareness of properties (p-awareness) and awareness of objects (o-awareness). He contrasts this with perception-that: awareness of facts (f-awareness). The latter, unlike the former, is concept (and belief) dependent. In the case of perception, we are o-aware, that is, object-aware, of perceptual objects: cats, motorcycles, scrambled eggs, and so forth. We are also p-aware of properties of those objects: their color, shape, smell, and so forth.11 One can simply see a black cat, be o-aware of the cat and p-aware of its blackness even if one does not possess the concepts of CAT or BLACK. That is, one can still see the black cat. One cannot see that the cat is black12, be f-aware that the cat is black, unless one sees the black cat and possesses (in some sense, see footnote 5) the concepts CAT and BLACK. (Additionally, seeing that the cat is black may depend on having certain background beliefs, e.g. believing that one is not hallucinating.) Thus (perceptual) f- awareness is dependent upon o- and p-awareness, the possession of certain concepts (O and P), and perhaps certain beliefs .

10 Audi (1998) calls these modes of awareness. 11 For present purposes, I will assume, on phenomenological grounds, that so-called secondary qualities: colors, tastes, smells, and so forth, are (physical) properties of the objects of perception. 12 Perception-that is often read as perceptual knowledge: Namely, if one sees that the cat is black, then it is true that the cat is black, and one believes that the cat is black. I am more inclined to characterize perception-that simply as belief. It is fallible and may conflict with other beliefs.

Psyche, Volume 16, number 1 Against treating introspection

This way of distinguishing different species of perception also provides a clear-cut way of describing cases of non-veridical perception. We might describe hallucination as property- awareness without object-awareness and illusion as object-awareness without (the relevant) property-awareness. While there is some debate as to whether we perceive objects in virtue of perceiving their properties, let us assume that in the case of veridical perception, o- and p- awareness go hand in hand. Seen in these terms, introspection differs from perception in that it does not seem that we are ever introspectively o-re or p-aware of mental objects (states) or their properties. It does seem, however, that we can be introspectively f-aware that our mental states have certain properties (that they are experiences of a black cat, for example). Moreover, it would seem that the phenomenal character of experience is a feature of simple perception (whether it is a non- intentional or an intentional property of simple perception we need not say here.) 13 Given that there does not seem to be anything like “simple introspection,” introspection unlike perception does not involve any phenomenal character over and above that of the state introspected. Moreover, if what is common to veridical and non-veridical perception is a similar phenomenal character, then introspection cannot be hallucinatory in this respect. Since what distinguishes veridical from non-veridical perception is the role of the perceptual object (in producing o- awareness perhaps via p-awareness), and since introspection does not seem to involve o- and p- awareness, some other explanation of the causal relation must be given. Finally, if introspection cannot be simple like perception, then it would seem that it is always f-awareness that is, in turn, concept dependent. In addition, whereas one can be aware of a black cat and not believe anything about black cats, and seeing that the cat is black seems to involve believing that the cat is black, fact-awareness is intricately tied to belief. If introspection is also a f-awareness, it too will be intricately tied to belief. Still, since a perceptual theory of introspection is one that likens introspection to perception on the basis that they share certain essential features, and introspection does not exhibit certain essential features of perception, we have reason to reject a perceptual theory of introspection. Perception is a causal relation that holds between a perceiver and a perceptual object (and its properties). In virtue of this perceptual awareness-of, or simple perception, the perceiver acquires information about the thing perceived such that, given the perceiver possesses the requisite concepts (and background beliefs), the perceiver can form beliefs (and acquire knowledge) about his or her environment. If one were not aware of perceptual objects (or their properties), then one could not be said to be perceiving them. In contrast, it seems that one can be introspectively aware that one is having an experience without being aware of the experience or its properties (which implies that if we

13 This can be seen by taking simple perception out of perception-that. For example, imagine being aware that the cat is black in virtue of being aware of the black cat. Now imagine being aware that the cat is black indirectly, for example if someone has told you her cat is black. Surely, in the former case there is something that it is like to be aware that the cat is black and this is not the case in the latter situation. Thus it seems that the phenomenal character of experience is a feature of simple perception rather than a feature of perception-that. Moreover, given the way I have described hallucinatory and illusory experience (above), we can say that there is reason to think that phenomenal character is a property of p-awareness since what is common to both veridical and non-veridical experience is p-awareness. To say this much is neutral with respect to the question of whether the phenomenal character is determined by qualia or by some intentional property of the awareness.

Psyche, Volume 16, number 1 Against treating introspection introspectively acquire information about the mental, then we do so only indirectly.) Additionally, whereas perceptual experience seems to involve there being an accompanying phenomenal character, introspection does not. Given this, while perception is subject to illusion (or hallucination) which is phenomenally indistinguishable from veridical perception, introspection is not subject to illusion or hallucination.14 Finally, whereas (simple) perception is not dependent upon possessing particular concepts and beliefs pertaining to the thing perceived, if introspection is a f-awareness, it is concept- and belief-dependent. The features of perception not found in introspection all appear to be features of simple perception, or awareness-of. This leaves open the possibility that introspection be construed as a species of perception-that, or awareness-that. This possibility will not be pursued here, however, though two things might be mentioned. First, in the case of perception, perception-that is dependent upon simple perception. Because introspection lacks anything analogous to simple perception, construing introspection as perception-that is insufficient to count as perceptual. Second, were one to argue that simple perception is not essential to perception, that all perception is perception-that (a species of belief) then perception does not meet the requirement of being basic. Thus, any perceptual theory of introspection that treats perception as belief will not count as a perceptual theory. For example, one might try to defend a perceptual theory of introspection based on an intentional theory of perception. According to an intentional theory of perception, perceiving (and perceptual experience15) is being in a state that has a particular content in virtue of representing the environment in a particular way. While an intentional theory of perception looks promising because it emphasizes certain similarities between introspection and perception, it does not, however, provide a basis for distinguishing perception from other intentional states, for example, belief. Therefore, it does not clearly satisfy the requirement that a perceptual theory of introspection be one for which perception is a basic mental state. If we accept that what is at the basis of a perceptual theory of introspection is the claim that there are certain essential features of perception that introspection shares, then right away we have reason to reject perceptual theories of introspection. We can admit that introspection and perception share some features: They are both mental states; they are both contentful; and they both provide us with information with which we can form beliefs and acquire knowledge. There are other features of perception, however, essential features, which introspection just does not exhibit. To deny that these are essential features of perception is just to be seriously confused about the role of perception. First, perception involves a sensation, appearance, or phenomenal character, and it is not clear that introspection does. We can, for present purposes, understand this phenomenal character simply as that which veridical and non-veridical perceptual experiences have in common. Such experiences are phenomenally indistinguishable. When one is (introspectively)

14 This is not to say that introspection is infallible. I simply mean to say that introspection lacks a defining characteristic of hallucination and illusion, namely, being phenomenally indistinguishable from veridical perception. 15 Note that I am approaching the rejection of a perceptual theory of introspection from the point of view of the qualia realist. Another option for the qualia realist is to reject any perceptual model of introspection that is inconsistent with qualia realism. An intentional theory of perception is not, on its own, inconsistent with qualia realism; however, an intentional theory of perception, coupled with an intentional theory of perceptual experience or phenomenal character, is inconsistent with qualia realism. 84

Psyche, Volume 16, number 1 Against treating introspection aware that one sees a black cat, however, the seeing itself does not appear as an object, nor does the introspective experience have any particular phenomenal character over and above the phenomenal character of the seeing state itself. That is, while there may be something that it is like to see a black cat, and surely there is, there is not some extra phenomenal quality associated with becoming introspectively aware of seeing the black cat. Since introspection lacks phenomenal character, it lacks an essential feature of perception. Second, there is nothing analogous to simple perception in the case of introspection. Simple perception occurs when one is aware of a perceptual object and its properties. In simply perceiving something, one need not possess any concepts of that thing (in simply seeing a black cat, one need not possess the concept CAT or BLACK.) Only in virtue of simply perceiving something can we perceive that it is so and so. Because we are not introspectively (o- or p-) aware of mental states, introspection, unlike perception, cannot be simple. Moreover, given this, the means to introspective beliefs (introspection-that) is just not the same as that to perceptual beliefs (perception-that).16 The perceptual belief that the cat is black is dependent upon being aware of the black cat. On the other hand, it seems that while we can (on the basis of introspection) believe certain things about our mental states, these beliefs are not dependent upon an awareness of these states or their properties. It is an essential feature of perception that there are these two species, or modes, of perceptual awareness that in the end allow for perceptual beliefs. Introspection lacks this feature of perception and so ought not to be conceived of on a perceptual model. Finally, while a perceptual theory of introspection grounded on an intentional theory of perception seems to emphasize certain similarities between perception and introspection (they are both states that represent the world in a certain way), this general similarity between introspection and perception is common to other states as well, for example, belief; therefore, it does not clearly take perception to be a basic mental state.

References

Armstrong, D. M. (1968). A materialist theory of the mind. New York: Routledge. Arnold, D. G. (1997). Introspection and its objects. Journal of Philosophical Research, 22, 87-94. Audi, R . (1988). Belief, justification, and knowledge. Belmont, CA: Wadsworth Publishing Company. Audi, R . (1998). Epistemology. New York: Routledge. Bealer, G . (1998). A theory of concepts and concept possession. Philosophical Issues, 9, 261-301. Crane, T . (1988). Concepts in perception. Analysis, 48(2), 150-153. Crane, T. (1992) The non-conceptual content of experience. In Crane (Ed.), The contents of experience (pp. 136- 57). Cambridge: Cambridge University Press..

16 Some beliefs may seem to be perceptual as they are about perceptual objects and yet are not (directly) caused by them. You might tell me the cat in the next room is black, and I might thereby come to believe the cat is black not ever having (simply) seen it. Beliefs about cats, in the usual sense, are described as perceptual because they are about the standard sort of perceptual objects; however, this sort of second-hand, or indirect, perceptual belief I will not count as perceptual per se unless it results from simple perception (in the same subject.)

Psyche, Volume 16, number 1 Against treating introspection

Dennett, D. (1991). Consciousness explained. Boston: Little, Brown and Company. Descartes, R. (1984). The philosophical writings of Descartes. In J. Cottingham, R. Stoothoff, & D. Murdoch (Trans.). Cambridge: Cambridge University Press. Dicker, G . (1980). Perceptual knowledge. Boston: D. Reidel Publishing Company. Dretske, F. (1981). Knowledge and the flow of information. Cambridge, MA: The MIT Press. Dretske, F. (1995). Naturalizing the mind. Cambridge, MA: The MIT Press. Dretske, F. (1999). The mind’s awareness of itself. Philosophical Studies, 95, 1-22. Dretske, F. (2000). Perception, knowledge and belief. Cambridge: Cambridge University Press. Evans, G . (1982). Varieties of reference. New York: Oxford University Press. Grice, H.P. (1961). A Causal Theory of Perception. Proceedings of the Aristotelian Society, Supplement Vol. 35, 121-152. Jackson, F. (1977). Perception: A representative theory. Cambridge: Cambridge University Press. Lycan, W. G. (1996). Consciousness and experience. Cambridge, MA: Bradford-MIT. Lyons, W. (1986). The disappearance of introspection. Cambridge, MA: The MIT Press. Macdonald, C. (1999). Shoemaker on self-knowledge and inner sense. Philosophy and Phenomenological Research, 59(3), 711-737. McDowell, J. (1994). The content of perceptual experience. The Philosophical Quarterly, 44(175), 190-205. Moore, G. E. (1922). Refutation of idealism. Philosophical Studies. London: Routledge & Kegan Paul. Originally published (1903). Mind, 7, 1-30. Peacocke, C . (1992). A study of concepts. Cambridge, MA: The MIT Press. Peacocke, C . (2001). Does perception have nonconceptual content? Journal of Philosophy, 98(5), 239-64. Searle, J. (1992). The rediscovery of the mind. Cambridge, MA: The MIT Press. Shoemaker, S. (1996). The first-person perspective and other essays. Cambridge: Cambridge University Press. Tye, M. (1995). Ten problems of consciousness: A representational theory of the mind. Cambridge, MA: Bradford- MIT. Tye, M. (2000). Consciousness, color, and content. Cambridge, MA: The MIT Press.